REVIEW

MitoTimer: a novel protein for monitoring mitochondrialturnover in the heart

Roberta A. Gottlieb & Aleksandr Stotland

Received: 7 September 2014 /Revised: 19 October 2014 /Accepted: 12 November 2014 /Published online: 6 December 2014# The Author(s) 2014. This article is published with open access at Springerlink.com

Abstract Mitochondrial quality control refers to the coordinat-ed cellular systems involved in maintaining a population ofhealthy mitochondria. In addition to mitochondrial proteinchaperones (Hsp10, Hsp60, and others) and proteases (Lon,AAA proteases) needed for refolding or degrading individualproteins, mitochondrial integrity is maintained through the reg-ulation of protein import via the TOM/TIM complex and pro-tein redistribution across the network via fusion and fission andthrough mitophagy and biogenesis, key determinants of mito-chondrial turnover. A growing number of studies point to theimportance of mitochondrial dynamics (fusion/fission) and mi-tochondrial autophagy in the heart. Mitochondrial biogenesismust keep pace with mitophagy in order to maintain a stablenumber of mitochondria. In this review, we will discuss the useof MitoTimer as a tool to monitor mitochondrial turnover.

Keywords MitoTimer .Mitochondrial biogenesis .

Mitophagy . Fluorescencemicroscopy

Mitochondrial turnover

Mitochondria have tissue-specific rates of turnover. For ex-ample, in the mouse heart, mitochondria turn over with a half-life of 14 days [1], but in the liver, the half-life is 2–4 days [2,3]. Selective mitochondrial autophagy, or mitophagy, elimi-nates damaged and dysfunctional mitochondria [4] and isclosely linked to mitochondrial biogenesis, which permitsreplacement of mitochondria (or synthesis of componentsand their insertion into the remaining functional mitochondria).

Rat cardiomyocytes have ~10,000 mitochondria per cell [5–9];this suggests that under resting conditions, one mitochondrialunit per cell is replaced every 4min (assuming turnover proceedsat a constant rate). Mitochondria comprise one third of thevolume of the heart and under normal conditions are responsiblefor 93% of ATP production (glycolysis is responsible for 3–7%,but this may rise considerably during ischemia) [10]. However,mitochondria are not producing ATP at maximal rates all of thetime, rather they are heterogeneous [11]: somemitochondriamaybe producing ATP at high rates while others may be relativelyinefficient or quiescent or producing significant amounts ofreactive oxygen species (ROS). Mitochondria routinely sustainoxidative damage as a by-product of oxidative phosphorylation.However, this is ordinarily handled by targeted protein qualitycontrol (Lon protease and other AAA proteases) [12] and byexchange of components during fusion followed by fission [1,11]. Asymmetric fission allows for segregation of dysfunctionalcomponents into one daughter mitochondrion that can be re-moved by autophagy [11]. A number of enzymes involved inoxidative phosphorylation (OX-PHOS) are expressed with diur-nal variation [13], suggesting that mitochondrial regeneration isregulated by the circadian cycle, and further implies that mito-chondrial elimination might be similarly regulated. This seemslikely given that a period of fasting occurs during sleep; therefore,waves of mitochondrial elimination may alternate with regener-ation on a daily basis, with 3–4 % of the mitochondrial mass (orpopulation) being replaced each day [3]. However, this may bemuch higher: in HL-1 cells subjected to serum and amino acidstarvation (glucose present) for 3.5 h, the number of mitochon-dria is reduced by 70 % [14].

Benefits of mitophagy

Previous work by our lab has demonstrated that Parkin-dependent mitophagy is essential for the protective effect of

R. A. Gottlieb (*) :A. StotlandHeart Institute and Barbra Streisand Women’s Heart Center,Cedars-Sinai Medical Center, 127 S. San Vicente Blvd. AHSP9105,Los Angeles, CA 90048, USAe-mail: roberta.gottlieb@cshs.org

J Mol Med (2015) 93:271–278DOI 10.1007/s00109-014-1230-6

ischemic preconditioning [15]. The selection of mitochondriafor elimination by autophagy is related to low membranepotential and possibly other factors such as ROS production[16]. Since mitochondria with low membrane potential willnot produce much ATP, their selective elimination may nothave a noticeable impact on cellular ATP levels. In fact,elimination of depolarized mitochondria could increase cellu-lar ATP by reducing the futile hydrolysis of ATP in an effort torestore mitochondrial membrane potential. Furthermore, elim-inating dysfunctional mitochondria might attenuate ROS pro-duction, and the remaining mitochondrial population mightexhibit better calcium homeostasis and greater resistance toMPTP induction. In rats subjected to 50 % caloric restrictionfor 1 week followed by 1 week of refeeding, the hepaticmitochondria were shown to be more efficient, as reflectedby higher state 3 mitochondrial respiratory capacity and in-creased superoxide dismutase activity [17].

Failure of mitochondrial quality control is accompanied byprogressively lower membrane potential, decreasing cellularATP levels, and increased ROS production [18]. Complex Iand aldehyde dehydrogenase 2 (ALDH2) are well known tobe vulnerable to oxidative damage, and specific aldehydemodifications are detected in mitochondria from hearts afteracute oxidative stress [19, 20]. Cardioprotection reduces theabundance of these modifications, and we suggest this ismediated in part by mitophagy and biogenesis.

Mitophagy is balanced by biogenesis

In order to maintain cellular homeostasis, elimination of mi-tochondria must be balanced by their replacement with newer,more efficient mitochondria. The transcriptional co-activatorperoxisome proliferator-activated receptor gamma co-activator 1-alpha (PGC-1α) is the master regulator of mito-chondrial biogenesis. Interestingly, many of the same regula-tory elements found in the promoters of OX-PHOS genes arealso found in the promoters of autophagy genes, suggestingcoordinate regulation (L van der Stap, KD Finley, and RAGottlieb, unpublished data). PGC-1α also regulates expres-sion of TFEB, the key transcriptional regulator of autophagyand lysosomal biogenesis. The peroxisome proliferator-activated receptor gamma (PPARγ) in tandem with theretinoic acid receptor (RXR) regulates pathways of fatty acidoxidation for both mitochondria and peroxisomes. Free fattyacids, clofibrate, and thiazolidinediones are ligands forPPARγ and st imulate mitochondrial biogenesis .Mitochondrial biogenesis is controlled by the PPARγ coacti-vator (PGC) family of transcriptional coactivators, most im-portantly PGC-1α, PGC-1β, and the PGC-related coactivatorPRC. PGC-1α works in tandem with nuclear respiratoryfactor 2 (NRF-2) to co-activate NRF-1 [21]. The NRFs directthe transcription of nuclear-encoded mitochondrial proteins,

the mitochondrial protein import machinery, and co-factorsrequired for assembly of the respiratory chain complexes, aswell as the regulatory factors required for mitochondrial DNAtranscription and translation, most importantly mitochondrialtranscription factor A (Tfam). Tfam is important not only formitochondrial gene transcription but also for maintenance ofmitochondrial DNA (mtDNA) copy number [22].Overexpression of PGC-1α is sufficient to drive mitochondri-al biogenesis [23]. Acetylation of PGC-1α suppresses itstranscriptional co-activator activity and can therefore limitmitochondrial biogenesis despite high levels of protein.Therefore, an important regulator of mitochondrial biogenesisis the histone deacetylase sirtuin 1 (Sirt1) which can serve toactivate autophagy as well as PGC-1α and mitochondrialbiogenesis [24, 25]. Another linkage between mitophagyand biogenesis is PARIS (also known as ZNF746), a substrateof Parkin that functions as a transcriptional repressor of PGC-1α [26].

Additional pathways converge on mitochondrial qualitycontrol

Mitochondria possess machinery for responding to an imbal-ance between imported nuclear-encoded proteins andmitochondrial-encoded proteins, known as the mitochondrialunfolded protein response (UPRmt). The correct stoichiome-try of OX-PHOS subunits is essential for assembly of func-tional respiratory complexes. Excess or unincorporated sub-units are degraded to short peptides by matrix proteases of theAAA+ family (ATPases associated with a variety of cellularactivities); Lon protease is responsible for the largest share ofintramitochondrial protein degradation. In Caenorhabditiselegans, the peptide fragments generated by the related prote-ase ClpP are extruded into the cytosol through the transporterHaf1 [27]. The mammalian homolog is thought to beABCme10, and the extruded peptides are recognized by tran-scription factors CHOP, C/EBPβ, and cJun/AP1, which directthe expression of mitochondrial chaperone hsp60 and prote-ases ClpP, Yme1L1, and MPPβ. Upregulation of these factorsreduces protein aggregation in the mitochondrial matrix.Additional proteins upregulated in the UPRmt includeTim17A, NDUFB2, and EndoG, but not ER stress proteinsBip, Grp94, calreticulin, or calnexin [28]. PKR (double-stranded RNA-activated protein kinase) phosphorylateseIF2α and cJun, with the resulting suppression of translationand import of nuclear-encoded mitochondrial proteins [29].There is almost nothing known about the significance of theUPRmt in the mammalian heart or its connection tomitophagy and mitochondrial biogenesis. However, agentssuch as chloramphenicol and rapamycin, which inducemitophagy/autophagy, are also reported to be potent inducersof UPRmt [30]. How the UPRmt may affect MitoTimer

272 J Mol Med (2015) 93:271–278

protein turnover and whether the overexpression ofMitoTimer affects the UPRmt are unknown but may need tobe considered when interpreting results.

Fluorescence properties of MitoTimer

Recent advances in assessing mitochondrial turnover includethe analysis of the half-lives of mitochondrial proteins usingmass spectrometry analysis and deuterium labeling [3], mito-Keima, which can report on mitochondria delivered to thelysosome [31], and MitoTimer, a fluorescent protein that canbe used to monitor mitochondrial turnover [32–34]. Kim et al.demonstrated that deuterium labeling coupled with gaschromatography-mass spectrometry can reveal differentialturnover rates of mitochondrial proteins on a proteome-widelevel [3]. This level of analysis of mitochondrial turnover rateis highly informative in whole organs but is not suitable foranalysis of mitochondrial turnover on a level of discrete

mitochondria. Likewise, while mito-Keima is a useful tool inquantitative assessment of single mitophagic events and al-lows for monitoring the rate of mitophagy in cells, it does notallow for monitoring of mitochondrial biogenesis. The majoradvantage provided by MitoTimer is that it allows for themonitoring of mitochondrial turnover and protein import(with resolution ranging from whole tissues down to individ-ual mitochondria). Timer is a fluorescent protein developed byTerskikh as a mutant of DsRed [35]. Timer molecules transi-tion from green fluorescence to a more stable red conforma-tion over a span of 48 h [Fig. 1]. Timer or fusion proteinderivatives have been used to monitor various processes in-cluding gene expression, intracellular protein recycling, cellsurvival, and the kinetics of viral infection [36–39]. For thepurpose of tracking mitochondrial turnover, a useful approachis to express MitoTimer under the control of a tetracyclineresponse element (TRE) such as pTRE-tight [33] in conjunc-tion with the reverse tetracycline transactivator (rtTA).Synchronized expression of MitoTimer is achieved by

Fig. 1 Maturation of MitoTimer in C2C12 cells. Cells stably transfectedwith rtTA and pTRE-tight-MitoTimer were exposed to doxycycline for1 h followed bymedia exchange, then imaged at the indicated times. Each

large panel shows the merged image while the insets show the individualred and green channels. Synchronized maturation of MitoTimer isevident

J Mol Med (2015) 93:271–278 273

inducing expression for a brief period of time with addition ofdoxycycline (Dox) pulse, followed by removal of any residualDox (chase). The protein can be monitored as it matures fromgreen to red over 48 h, and it is retained in the mitochondriafor a number of days (presumably corresponding to a mito-chondrial half-life). A second Dox pulse can be used to induceanother round of expression, which is readily evident as greenMitoTimer [Fig. 2 and schematic shown in Fig. 3]. MitoTimermatures within 48 h, a rate which parallels mitochondrialturnover in the liver, but considerably faster than turnoverrates in the heart. Thus in tissues with slow mitochondrialturnover, the presence of red mitochondria after a brief induc-tionwith doxycycline cannot allow one to distinguish betweenmitochondria that imported MitoTimer 2 days ago and mito-chondria that imported the protein 1 or 2 weeks ago. Toexamine mitophagy, one must monitor the rate of disappear-ance of (red) MitoTimer, but in dividing cells (even stablytransfected cells), one cannot distinguish between dilution ofthe existing fluorescent protein due to cell division and loss ofprotein through mitophagy. Thus monitoring basal rates ofmitochondrial turnover in immortalized cells may be difficult.However, MitoTimer can be used tomonitor mitophagy that isacutely upregulated in response to a stressor. Similar chal-lenges and limitations exist with monitoring mitochondrialbiogenesis, but these can be partially overcome by using a

second pulse of doxycycline before or during the interventionthat is proposed to induce biogenesis. A diagram of the typicaltwo-pulse labeling protocol is shown in Fig. 3.While a seconddoxycycline addition triggers a rather small amount ofMitoTimer incorporation under basal conditions (probably3–4%, corresponding to the net turnover rate), an interventionthat triggers biogenesis will result in a much larger incorpo-ration of newly synthesizedMitoTimer. Given that MitoTimeris under the control of the artificial TRE-tight promoter, thedifferential protein translation and import under basaland stimulated conditions is likely due to posttranscrip-tional control through the native cox8 mitochondrialtargeting sequence added to Timer [40]. Finally, it ispossible to gain insight into the relative equilibrium ofmitophagy and biogenesis under conditions of constitu-tively expressed MitoTimer, where a change in the redto green ratio reflects a shift in the equilibrium.

Specific considerations and protocols for usingMitoTimer

When utilizing MitoTimer to monitor mitochondrial turnover,several factors need to be taken into account. Timermaturation rate has previously been shown as independentof protein concentration, ionic strength, and a wide range of

Fig. 2 Protein import in the mitochondrial network is spatiallyinhomogeneous. A field of C2C12 cells exposed to doxycyclinecontinuously for 24 h is shown. Rectangles outline areas that are

enlarged at right; arrows indicate regions of the mitochondrial networkwhich contain green MitoTimer (newly synthesized and imported) butlack mature (red) MitoTimer

274 J Mol Med (2015) 93:271–278

biological pH values [35]. In our studies with MitoTimer, wehave found that the maturation rate was, however, acceleratedduring live fluorescence imaging of the cells. When fixed in4 % paraformaldehyde, MitoTimer retains the conformation itwas in at the time of fixation, thus permitting prolongedimaging of the cells. Furthermore, depending on the fluores-cent microscope used in the experiments, filter sets may needto be optimized in order to detect the green signal ofMitoTimer. For example, in our previous studies, we utilizeda Nikon (TE300, Nikon, Melville, NY) microscope equippedwith a cooled CCD camera (Orca-ER, Hamamatsu,Bridgewater, NJ) and automated excitation emission filterwheels controlled by a Lambda 10–2 (Sutter instruments)operated by MetaMorph Version 6.2r (Molecular Devices);in order to maximize the MitoTimer signal, we used a triple-cube Chroma 61002-triple emitter 61002, with the addedgreen emission filter Chroma 233755 D520–40 m (midpoint520, emission 500–540) and the red emission filter Chroma221379D605–55 (midpoint 605, emission 578–632). It is alsoimportant to make sure that threshold levels used in collectingthe images from MitoTimer cells remain the same throughoutthe time course experiment. Inducible expression ofMitoTimer is a useful approach in experiments utilizing celllines, as time-restricted induction can provide insights intomitochondrial dynamics and subpopulation. Inducible expres-sion in a stably transfected cell line also avoids the effect ofleaky plasmid promoters, heterogeneous protein expressionintrinsic to transient transfection, and the abnormal

accumulation of messenger RNA (mRNA), all of which maymask the visualization of MitoTimer maturation in discretemitochondria. Constitutive expression of MitoTimer can,however, be informative in in vivo experiments, as demon-strated by Laker et al. [32]. In this context, mitochondrialdynamics and import rates are much more heterogeneous thanthose observed in immortalized tumor-derived cell lines (es-pecially if expressed under tissue-specific promoters), and ashift in the abundance of newly expressed vs. matureMitoTimer can be used to reveal changes in the rate of mito-chondrial turnover (the balance of biogenesis and mitophagy).In our work, we utilized the well-characterized tetracycline(Tet)-On inducible expression systemwith a brief doxycycline(Dox) pulse in order to synchronize the uptake of newlysynthesized MitoTimer by the mitochondria. Although thissystem allows for conditional expression, it still has the draw-back of requiring transient transfection with two plasmids andcan be unwieldy given the need to transfect, induce expres-sion, and monitor maturation of MitoTimer during a particularintervention. To solve this problem, we found that the bestlevel of control in the system could be achieved by usingretroviral transfer vectors to express the reverse Tettransactivator and MitoTimer under the control of the Tetresponsive elements, which we used to establish stable celllines derived from the C2C12 skeletal muscle myoblast cellline and the murine atrial cardiomyocyte HL-1 cell line. Wefurther improved the control of the system using fluorescence-activated cell sorting to sort first for cells that were negative

2h after dox #2 with minimal biogenesis 6-12h after dox #2 with minimal biogenesis

dox pulse #1 6hr 12hr 24hr 48hr and longer dox pulse #2

2h after dox #2 with active biogenesis 6-12h after dox #2 with active biogenesis

Fig. 3 Schematic of two-pulseDox induction ofMitoTimer. Dox is added tomedia for 2 h, then washed out, andMitoTimer is allowed to mature for 48 h,after which a second pulse of Dox is given in the presence or absence of anagent to stimulate mitochondrial biogenesis. Cells are imaged a few hours

later. Under conditions of active mitochondrial biogenesis, more mitochon-dria have imported new (green) MitoTimer than in unstimulated cells, wherethere is minimal biogenesis and correspondingly little import of newlysynthesized MitoTimer protein

J Mol Med (2015) 93:271–278 275

for MitoTimer in the absence of Dox; the sorted cells werethen stimulated with Dox followed by a second round ofsorting to select cells with the highest expression ofMitoTimer. The resulting cell lines exhibited tight control ofMitoTimer expression and required only a brief (2 h) pulsewith 2 mg/ml Dox in order to achieve a high level of expres-sion. Following the pulse, it is important to completely re-move Dox from the media by washing several times withphosphate-buffered saline and replacing with fresh media(and tetracycline-free serum), as even a trivial concentrationof Dox is sufficient to induce expression in cell lines selectedfor high responsiveness. Utilizing the method of cell lineselection described above, we have been able to expressMitoTimer in an inducible fashion across a wide range of celllines; this approach is strongly recommended for in vitrostudies. MitoTimer is also a tool highly suitable for analyzingand sorting mitochondria by flow cytometry. When used forthis purpose, it is highly recommended that single positivegreen fluorescent and red fluorescent cells or mitochondria beused in setting up the parameters of the sort to compensate forthe bleed-through of green fluorescence into the red channel.This step is crucial in obtaining an accurate picture of thelevels of green-to-red fluorescence of individual cells andmitochondria and should be performed alongside theMitoTimer experiments as part of the controls.

Use of MitoTimer to investigate mitochondrial biology

We targeted Timer to the mitochondrial matrix (MitoTimer).The rate of diffusional exchange of contents across the mito-chondrial network is fastest for outer membrane constituents,intermediate for matrix contents, and slowest for inner mem-brane proteins [18]. Mitochondrial fusion events require highmembrane potential [11] and mitofusin-1 or mitofusin-2(Mfn1/2) [34]. In cells with a normal mitochondrial network,MitoTimer of varying degrees of maturation (green to red) isdistributed homogeneously across the network. However, inmouse embryonic fibroblasts (MEFs) derived from Mfn1/2double knockout mice, distribution of new vs. matureMitoTimer was heterogeneous [34], demonstrating that fusionevents enable mixing of old and new proteins within themitochondrial network. Neurons also exhibit heterogeneousdistribution of MitoTimer, with newly synthesized proteinbeing incorporated into mitochondria closest to the somawhile mature MitoTimer predominated in the distal portionof the neurites. In fact, the red to green ratio was proportionalto distance from the soma [34]. Interestingly, it has beenreported that autophagosomes initiate distally in neurons (nearthe oldest mitochondria) and then migrate towards the soma,where they fuse with lysosomes [41]. These examples illus-trate the use of MitoTimer to gain spatiotemporal informationabout mitochondrial age.

Use of MitoTimer to assess protein import

MitoTimer also provides information about the level of mito-chondrial protein import across a population of cells andorganelles. High membrane potential is required for mito-chondrial protein import [42]. Mitochondria which lose mem-brane potential will be unable to import new (green)MitoTimer protein. If biogenesis is impaired, one would ex-pect a decrease in the number of mitochondria importing newprotein, which would be reflected by a decrease in greenMitoTimer in mitochondria. Conversely, increased biogenesiswould be reflected in an increase in green MitoTimer inmitochondria. This was demonstrated in our study in whichwe showed that cells recovering from FCCP or statin treat-ment showed an increase in new MitoTimer protein import(green mitochondria) [33] during a second period of doxycy-cline induction (using a tetracycline-inducible construct andrtTA-expressing cells). Using constitutive expression ofMitoTimer in a Drosophila heart tube, Laker et al. showedthat exercise training resulted in an increase in the abundanceof green mitochondria [32], consistent with increased mito-chondrial biogenesis. Oddly, the authors attributed this greenpredominance to decreased oxidative stress in the exercisedmice, although it has been shown that exercise increasesreactive oxygen species in skeletal muscle mitochondria[43], and MitoTimer color maturation was shown to be insen-sitive to reactive oxygen species [34]. Interestingly,MitoTimer import during biogenesis is not uniform but isenriched in a perinuclear zone in C2C12 cells [Fig. 2] [33]and in neuronal cells [34]. While this may be simply due tomRNA proximity, it is possible that the subpopulation ofmitochondria enriched for green MitoTimer is specializedfor mitochondrial regeneration.

Use of MitoTimer to assess mitochondrial turnover

Mitochondrial turnover also depends upon clearance of mito-chondria by mitochondrial autophagy or mitophagy.Suppression of autophagy results in the accumulation of mi-tochondria with lower membrane potential and increased ox-idative damage [44]. In autophagy-deficient cells in whichMitoTimer expression was induced for 4 h and allowed tomature for 48 h, there was increased accumulation of redMitoTimer compared to autophagy-competent cells [34], in-dicating that mitochondrial turnover was impaired. Similarly,in mice in which MitoTimer was constitutively expressed inskeletal muscle (after electroporation of virus), the ratio of redto green MitoTimer was increased in mice on a high-fat dietcompared to normal chow-fed animals; the ratio was de-creased in mice subjected to an exercise regimen comparedto sedentary mice on the same diet [32]. Collectively, thesestudies show that MitoTimer can be used to assess

276 J Mol Med (2015) 93:271–278

mitochondrial turnover. Analysis of mitochondrial turnoverwill contribute to our understanding of disease as acquiredmitochondrial dysfunction can be compensated for throughmitophagy and biogenesis. Loss of this compensatory processresults in the accumulation of damaged mitochondria whichmay generate increased reactive oxygen species; consumeATP through reversal of the ATP synthase; disrupt calciumhomeostasis; activate innate immunity, leading to inflamma-tion; or release cytochrome c, thereby triggering cell death.

Caveats and limitations of MitoTimer

The use of MitoTimer to date has been limited to cell culture,with limited studies in flies and electroporation into mouseskeletal muscle. However, the findings have engendered con-siderable enthusiasm for the technology, and the pTRE-tight-MitoTimer plasmid has been requested by 18 groups since itwas deposited with Addgene in April. There are some caveatsto consider when using MitoTimer, however. Given the widerange of turnover of different proteins in mitochondria [3], it isimportant to confirm that MitoTimer turnover is reflectingmitochondrial turnover. This can be done by testing interven-tions that either inhibit or accelerate mitophagy to confirm thatMitoTimer clearance changes accordingly. Similar verifica-tion is needed for estimates of biogenesis. Unfortunately, thechemical treatment used for 5-ethynyl-2′-deoxyuridine (EdU)staining destroys fluorescent proteins, making the two mutu-ally incompatible, although paired samples can be stained.Nevertheless, if both protocols show the same trend, thenone can feel assured that MitoTimer is behaving as one mightexpect, with respect to biogenesis. The fluorescent protein issubject to photoconversion (green to red) by persistent lightexposure, so time-lapse imaging may not be feasible. Use inprimary cells depends upon having the appropriate viral vec-tors for efficient gene delivery, and establishing transgenicmice will be invaluable.

Comparison of methods to monitor mitophagy

Currently three recently introduced methods provide informa-tion on mitochondrial turnover. The use of deuterium labelingand mitochondrial proteome analysis is extremely powerfuland can be used in cells, non-transgenic animals, and humans[3]. However, it provides limited subcellular information (al-though the interfibrillar and subsarcolemmal mitochondrialproteomes could be measured separately) and requires sophis-ticated mass spectrometry and bioinformatics capability.Mito-Keima is an elegant reporter for mitophagy [31] thatwill become extremely useful once transgenic lines are avail-able. Its chief limitation is that it can only report on the finalphase of mitophagy (delivery to the lysosome) and is only

applicable for fluorescence microscopy. MitoTimer providesinformation on biogenesis and, by using time-restricted ex-pression, can also provide information on mitophagy [33].Constitutive expression of MitoTimer can provide informa-tion on the relative balance of mitophagy and biogenesis [32,34]. Currently the approach is limited to cell culture or elec-troporation into tissues; however, adenoviral delivery to theheart is feasible, as is production of a transgenic mouse witheither the tetracycline-inducible promoter or a constitutivepromoter. Mitochondrial biogenesis can also be monitored incells or in vivo by incorporation of tagged nucleotides such asBrdU or EdU into mitochondrial DNA; successful labeling ofmtDNAwas described by Oka et al. [45].

Conclusions

Impaired mitochondrial quality control is frequently a conse-quence of impaired autophagy in conditions such as advancedage or the metabolic syndrome. Loss of mitochondrial qualitycontrol may contribute to many chronic disease phenotypes. Theuse of MitoTimer to monitor mitochondrial turnover may revealnew insights into mitochondrial biology in health and disease.

Open Access This article is distributed under the terms of the CreativeCommons Attribution License which permits any use, distribution, andreproduction in any medium, provided the original author(s) and thesource are credited.

References

1. Menzies RA, Gold PH (1971) The turnover of mitochondria in avariety of tissues of young adult and aged rats. J Biol Chem 246(8):2425–2429

2. Miwa S, Lawless C, Von Zglinicki T (2008) Mitochondrial turnoverin liver is fast in vivo and is accelerated by dietary restriction:application of a simple dynamic model. Aging Cell 7(6):920–923

3. Kim TY et al (2012) Metabolic labeling reveals proteome dynamicsof mouse mitochondria. Mol Cell Proteomics 11(12):1586–94

4. Kim I, Lemasters JJ (2010) Mitophagy selectively degrades individ-ual damaged mitochondria after photoirradiation. Antioxid RedoxSignal 14(10):1919–1928

5. Page E, McCallister LP, Power B (1971) Sterological measurementsof cardiac ultrastructures implicated in excitation-contraction cou-pling. Proc Natl Acad Sci U S A 68(7):1465–1466

6. McDonald K et al (1997) The effect of delayed reperfusion followinginfarction in the rat on structural changes in viable myocardium.Cardiovasc Res 36(3):347–353

7. Page E, McCallister LP, Power B (1971) Stereological measurementsof cardiac ultrastructures implicated in excitation-contraction cou-pling. Proc Natl Acad Sci 68(7):1465–1466

8. Fawcett DW, McNutt NS (1969) The ultrastructure of the cat myo-cardium. I. Ventricular papillary muscle. J Cell Biol 42(1):1–45

9. Nouette-Gaulain K, et al. (2005) Time course of differential mito-chondrial energy metabolism adaptation to chronic hypoxia in rightand left ventricles Cardiovasc Res. 66(1):132-40.

10. Ingwall JS (2002) ATP and the heart. Kluwer Academic Publishers,Norwell

J Mol Med (2015) 93:271–278 277

11. Twig G et al (2008) Fission and selective fusion govern mitochondrialsegregation and elimination by autophagy. EMBO J 27(2):433–446

12. Bender T, Lewrenz I, Franken S, Baitzel C, Voos W (2011)Mitochondrial enzymes are protected from stress-induced aggrega-tion by mitochondrial chaperones and the Pim1/LON protease. MolBiol Cell 22(5):541–554

13. Bailey SM, Udoh US, Young ME (2014) Circadian regulation ofmetabolism. J Endocrinol 222(2):R75–96

14. Carreira RS, Lee Y, Ghochani M, Gustafsson AB, Gottlieb RA(2010) Cyclophilin D is required for mitochondrial removal byautophagy in cardiac cells. Autophagy 6(4):462–72

15. Huang C et al (2011) Preconditioning involves selective mitophagymediated by Parkin and p62/SQSTM1. PLoS One 6(6):e20975

16. Kim I, Lemasters JJ (2011) Mitophagy selectively degrades individ-ual damaged mitochondria after photoirradiation. Antioxid RedoxSignal 14(10):1919–1928

17. Crescenzo R et al (2006) Altered skeletal muscle subsarcolemmalmitochondrial compartment during catch-up fat after caloric restric-tion. Diabetes 55(8):2286–2293

18. Wikstrom JD, Twig G, Shirihai OS (2009) What can mitochondrialheterogeneity tell us about mitochondrial dynamics and autophagy?Int J Biochem Cell Biol 41(10):1914–1927

19. Moon K-H, Kim B-J, Song BJ (2005) Inhibition of mitochondrialaldehyde dehydrogenase by nitric oxide-mediated S-nitrosylation.FEBS Lett 579(27):6115–6120

20. Paradies G, Petrosillo G, Pistolese M, Ruggiero FM (2002) Reactiveoxygen species affect mitochondrial electron transport complex Iactivity through oxidative cardiolipin damage. Gene 286(1):135–141

21. Lehman JJ et al (2000) Peroxisome proliferator-activated receptorgamma coactivator-1 promotes cardiac mitochondrial biogenesis. JClin Invest JID 7802877 106(7):847–856

22. Scarpulla RC (2008) Transcriptional paradigms in mammalian mito-chondrial biogenesis and function. Physiol Rev 88(2):611–638

23. Wende AR et al (2007) A role for the transcriptional coactivatorPGC-1α in muscle refueling. J Biol Chem 282(50):36642–36651

24. Iwabu M et al (2010) Adiponectin and AdipoR1 regulate PGC-1alpha and mitochondria by Ca (2+) and AMPK/SIRT1. Nature464(7293):1313–1319

25. Sengupta A, Molkentin JD, Yutzey KE (2009) FoxO transcriptionfactors promote autophagy in cardiomyocytes. J Biol Chem 284(41):28319–28331

26. Shin JH et al (2011) PARIS (ZNF746) repression of PGC-1alphacontributes to neurodegeneration in Parkinson’s disease. Cell 144(5):689–702

27. Haynes CM, Yang Y, Blais SP, Neubert TA, Ron D (2010) The matrixpeptide exporter HAF-1 signals a mitochondrial UPR by activating thetranscription factor ZC376.7 in C. elegans. Mol Cell 37(4):529–540

28. Aldridge JE, Horibe T, Hoogenraad NJ (2007) Discovery of genesactivated by the mitochondrial unfolded protein response (mtUPR)and cognate promoter elements. PLoS One 2(9):e874

29. Haynes CM, Ron D (2010) The mitochondrial UPR—protecting organelle protein homeostasis. J Cell Sci 123(Pt22):3849–3855

30. Mottis A, Jovaisaite V, Auwerx J (2014) The mitochondrialunfolded protein response in mammalian physiology. Genome,Mamm

31. Katayama H, Kogure T, Mizushima N, Yoshimori T,Miyawaki A (2011) A sensitive and quantitative techniquefor detecting autophagic events based on lysosomal delivery.Chem Biol 18(8):1042–1052

32. Laker RC et al (2014) A novel MitoTimer reporter gene for mito-chondrial content, structure, stress, and damage in vivo. J Biol Chem289(17):12005–12015

33. Hernandez G et al (2013) MitoTimer: a novel tool for monitoringmitochondrial turnover. Autophagy 9(11):1852–1861

34. Ferree AW et al (2013) MitoTimer probe reveals the impact ofautophagy, fusion, and motility on subcellular distribution of youngand old mitochondrial protein and on relative mitochondrial proteinage. Autophagy 9(11):1887–1896

35. Terskikh A et al (2000) “Fluorescent timer”: protein that changescolor with time. Science 290(5496):1585–1588

36. Loson OC, Ha CM, Parpura V (2011) Age-dependent spatial segre-gation of synaptobrevin 2-containing vesicles in astrocytes. JNeurochem 116(5):909–15

37. Kozel BA et al (2006) Elastic fiber formation: a dynamic view ofextracellular matrix assembly using timer reporters. J Cell Physiol207(1):87–96

38. Verkhusha VV et al (2001) An enhanced mutant of red fluorescentprotein DsRed for double labeling and developmental timer of neuralfiber bundle formation. J Biol Chem 276(32):29621–29624

39. Robinson SM et al (2014) Coxsackie virus B exits the host cell inshed microvesicles displaying autophagosomal markers. PLoSPathog 10(4):e1004045

40. Weis BL, Schleiff E, ZergesW (2013) Protein targeting to subcellularorganelles via mRNA localization. Biochim Biophys Acta 1833(2):260–273

41. Maday S,Wallace KE, Holzbaur EL (2012) Autophagosomes initiatedistally and mature during transport toward the cell soma in primaryneurons. J Cell Biol 196(4):407–417

42. Neupert W, Herrmann JM (2007) Translocation of proteins intomitochondria. Annu Rev Biochem 76(1):723–749

43. Aoi W, Naito Y, Yoshikawa T (2014) Potential role of oxidativeprotein modification in energy metabolism in exercise. SubcellBiochem 77:175–187

44. Wikstrom JD et al (2007) {Beta}-Cell mitochondria exhibit mem-brane potential heterogeneity that Can Be altered by stimulatory ortoxic fuel levels. Diabetes 56(10):2569–2578

45. Oka T et al (2012) Mitochondrial DNA that escapes fromautophagy causes inflammation and heart failure. Nature485(7397):251–255

278 J Mol Med (2015) 93:271–278